Hi,
i use Eigen 2.0, i have a generalized eigenvalue problem to solve (with general matrix), i know about the Generalized SelfAdjoint EigenSolver but it works only for selfadjoint matrix ..
does any other method exist in Eigen 2.0 or other versions to solve the generalized eigenvalue problem with general matrix ?
thanks for your help and forgive me my mistakes

There is a class EigenSolver, see the documentation:

http://eigen.tuxfamily.org/dox/classEig ... olver.html

There is no generalized eigensolver for non selfadjoint matrices. It is not as easy than for the selfadjoint case because it requires special versions of the Schur dec., and special Hessenberg dec.... I doubt we will propose one soon, unless someone step by.

The class EigenSolver is used for solving the eigenvalue problem : Ax=(lambda)x.
I search for a method to solve the generalized eigenvalue problem:
Ax=(lambda)Bx ; with A,B general matrix

the class GeneralizedSelfAdjointEigenSolver solves the problem with:
A :Selfadjoint matrix and B:Positive-definite matrix

but, i need a solution for general matrix ,(the algorithm QZ solves this problem using a generalized Schur decomposition) the function exist in lapack

thank you for help

ggael wrote:There is no generalized eigensolver for non selfadjoint matrices. It is not as easy than for the selfadjoint case because it requires special versions of the Schur dec., and special Hessenberg dec.... I doubt we will propose one soon, unless someone step by.

thank you, I wrote my reply to "phr" before seeing yours

oh, right

you can try interfacing with lapack directly, see

http://eigen.tuxfamily.org/index.php?title=Lapack

the example used on that page is in fact solving generalized eigenvalue problems using DGGEV

edit: but I don't understand the EigenToFortran and FortranToEigen methods on that page, why not use just m.data()?